How DeepSeek is Accelerating Llama: A Social Media Story from Zuckerberg’s Second Day at the Glimpse
Zuckerberg noted that “there’s a number of novel things they did we’re still digesting” and that Meta plans to implement DeepSeek’s “advancements” into Llama. DeepSeek caused a massive sell-off in AI stocks due to fears that models will no longer need as much computing power. With all of the billions of dollars he’s investing in Graphics Processing Unit, I think that it’s a strategic advantage over time.
Zuckerberg also made sure to praise President Donald Trump. He said the US administration is proud of its leading companies and will defend its values and interests abroad. Moments before the earnings call started, news broke that Meta is paying Trump $25 million to settle a lawsuit he brought against the company for banning his account after the January 6th insurrection. (The vast majority of the money is going to pay for Trump’s presidential library.)
Meanwhile, Meta is a cash-printing machine. Revenue for the fourth quarter of 2024 was $48.39 billion — a 22-percent increase from the year-ago period — while net profit was a staggering $20.8 billion (up 43-percent from a year before). During the earnings call, the CFO said that Meta hasn’t seen any noticeable impact from its content policy changes. 3.35 billion people used at least one of Meta’s apps daily in the fourth quarter — a 5-percent increase from the year-ago period.
DeepSeek and Stargate: What’s Happening in AI Armed Systems and Where the AI Sector is Going Next? Kylie Robison on Verge
DeepSeek is much like a mobile app and website, you can type into a text box to have it talk to you, and it will talk back to you. What makes it special is how it was built. The most recent major release by the startup was called R1, which was released just a few weeks after the last model V3 began showing some impressive benchmark performance. It quickly became clear that DeepSeek’s models perform at the same level, or in some cases even better, as competing ones from OpenAI, Meta, and Google. They are completely free to use.
DeepSeek claims that it cost less than $6 million to train, despite the fact that Open Artificial Intelligence reported training cost of as much as $100 million. In a matter of days, DeepSeek went viral, becoming the No. 1 app in the US, and on Monday morning, it punched a hole in the stock market.
More than $1 trillion of tech stocks were wiped off in a selloff earlier this week. Nvidia, in particular, suffered a record stock market decline of nearly $600 billion when it dropped 17 percent on Monday.
Tech executives have been telling us for more than two years that the path to unlocking the full potential of neural networks was to throw graphics cards at the problem. Since then, scale has been in charge. And scale was certainly top of mind less than two weeks ago, when Sam Altman went to the White House and announced a new $500 billion data center venture called Stargate that will supposedly supercharge OpenAI’s ability to train and deploy new models.
What is clear is that we’ve entered a new phase in the AI arms race, and DeepSeek and Stargate represent more than just two distinct paths toward superintelligence: they also represent a new, escalating front in the US-China relationship and the geopolitics of AI. This is getting more difficult as President Trump continues to damage foreign relations with a new threat of tariffs.
There is a whole lot going on here — and the news cycle is moving very fast. So to break it all down, I invited Verge senior AI reporter Kylie Robison on the show to discuss all the events of the past couple weeks and to figure out where the AI industry is headed next.